The news media love reporting on AI as if we are on the brink of machines taking over most traditional human jobs, such as writers, mechanics, doctors, lawyers, software developers, etc - any profession, really, that requires a good amount of knowledge, intuition and creativity. These headlines do catch attention and sell well, but how much intelligence really is there behind that blinking cursor of an AI prompt?

In the good old C++, types are distinct based on just their names, so one could have two empty classes A and B and instances of each could not be mixed up in function calls expecting one type or another. A template class instantiated with two different types, such as X<A> and X<B>, would also yield two distinct template instantiations that could not be mixed up in function calls and assignments.

Nowadays, structural typing in TypeScript focuses on what the type contains, and the type name doesn't play as much of a role in assignments. This logic extends to type parameters, which only have meaning when they are referenced in specific ways in their contexts.

This arrangement makes it quite tricky to figure out some of the typing errors, which sometimes sound more like puzzles rather than meaningful errors.

One of the recent Windows 11 updates described in release notes that passkeys are now easier than ever to manage in Windows Settings, so I gave it a go to see if it is time to go passwordless for websites that support Web Authentication.

After poking around this functionality for a bit, however, I realized that the line between going passwordless and ending up passwordless is not as bold as one would hope with passkeys, because this technology is not well documented, relies on data hard-wired into a specific device, and on how well tech behemoths, such as Microsoft, Google and Apple, interact with one another, which is not the strongest trait for many of them.